Why More Customers Don't Always Create More Value

UPS deliberately gave up package volume from its largest customer.

Its 2025 Form 10-K describes a plan to reduce Amazon volume by more than 50% by June 2026 relative to 2024. Amazon and affiliates accounted for 10.6% of 2025 revenue. Average daily package volume declined 7.0%, while revenue per piece rose 6.6%. UPS's operating profit also fell in 2025, so this shows a deliberate choice, not yet a result.

What exactly does the next customer cause the company to do?

A large account may bring attractive revenue alongside unusual inventory, engineering changes or special delivery. Another customer may pay a premium for genuinely demanding service. A third may look difficult because the company makes ordinary work unnecessarily complicated. These are different management decisions.

Not every dollar of revenue deserves the same operating capacity.

Four public-company disclosures illustrate customer selection, alternative delivery arrangements, historical costs from customer-directed changes, and customization that can command higher margins. None establishes a company-specific amount of Structural Drag.

The purpose is to distinguish what the customer requires, what the customer pays for, and what the company's arrangement makes people do to deliver it.

1. The Customer-Arrangement Map

Customer profitability analysis is well established. Pricing, margin and cost-to-serve matter, but those measures may stop one question too early.

Two companies may fulfill the same promise at the same price while using different amounts of paid effort. Scale, geography, skill and capacity may explain the difference; so may the arrangement of work.

In my Business Structural Drag framework, Structural Drag is work a company repeats because of how it is arranged—work a different arrangement could remove without changing what the customer receives or what the business must protect.

The analysis must therefore separate the commercial proposition from the operating design.

An expensive customer may require expensive service, pay too little for it, or trigger work that a different arrangement could avoid.

The customer is not the diagnosis. The arrangement is the hypothesis.

Consider a fictional industrial distributor promising a contractor next-morning parts while accepting late changes. Sales, purchasing, warehouse and dispatch use different release schedules, so staff repeatedly repick and expedite the same orders. The customer may legitimately need late changes and be willing to pay for them. But management should test whether the repeated work is necessary or whether compatible release rules would preserve the promise at lower cost. Repricing and redesign might both be warranted. This is an illustration, not a reported engagement.

2. The four signatures of arrangement-created work

The distinctive Structural Drag question starts with the difference between work necessary to keep a promise and work repeated because of the company's arrangement. Four signatures matter:

These signatures suggest structural hypotheses, not diagnoses. The six profiles are illustrated in Research 01.

The customer may have requested the exception. The company may have created the repetition.

Premium service may deserve premium pricing. Internal repetition may deserve redesign. Neither conclusion follows from gross margin alone.

3. Four explanations before calling a customer unprofitable

The same low contribution can have different causes. Test competing explanations before changing prices, refusing an account or redesigning work.

An allocated fixed cost may not disappear when an account leaves. Conversely, a customer may cover direct costs while consuming scarce capacity needed for better orders. Test incremental economics rather than declaring either case proven from allocations.

4. Four public-company situations

These are four documented situations, not Structural Drag diagnoses. The structural questions are mine.

5. What the four situations establish

UPS shows revenue selection; Coca-Cola Consolidated shows a choice of delivery route; Celestica documents costs from customer-directed changes; Builders FirstSource shows why valuable customization must be protected. None is a measured Structural Drag case.

Before calling a customer good or bad, establish what the customer requires, what the company earns, and what the operating arrangement makes everyone do.

A demanding account may deserve a premium; a low-margin account may be easy to serve; an apparently difficult account may become attractive if repeated internal work is removed. Match the decision to the evidence.

6. The economics: contribution, capacity and cash

Customer-level financial analysis must be sufficiently detailed to identify different decisions, but not so elaborate that arbitrary allocations determine which accounts look profitable.

The Customer Arrangement Bridge follows:

Net customer revenue → direct product and service costs → customer-specific incremental operating costs → economically relevant capacity and capital requirements → contribution and cash.

Reconcile actual prices, discounts and returns; direct fulfillment expenses; and observed incremental costs such as special inventory, repeat picking, engineering changes, exception handling or rework.

Then test scarce capacity and capital requirements. Opportunity cost matters when a customer prevents higher-contribution work—but it is not a cash expense, and cannot be counted again as a realized saving.

Separate incremental costs from common-cost allocations. A warehouse lease may remain when a customer leaves, but a service pattern may trigger the next shift, building or supervisor. Volume and timing matter.

My three economic ledgers remain distinct: verified financial effects, paid capacity consumed, and contribution not earned. Do not add a potential opportunity to realized savings or count released hours twice.

Payment terms and customer inventory also affect cash (Research 03); changing a service route does not automatically raise throughput (Research 04).

The decision is whether to retain, reprice, serve differently or decline the work—on incremental economics and observed operations.

A margin report tells you what the account appears to earn. A work sample tells you what the company must do to earn it.

7. Give this to your CFO, COO and commercial leader

The Customer Arrangement Test

Do not begin with a ranking of the ten least profitable customers. Begin with one consequential account or customer type whose revenue, growth, service complexity or operating demands could change the next capital or hiring decision.

1. What did we actually promise? Recover pricing, scope, volume, customization, delivery, service levels, returns, payment terms and exceptional rights. Identify what was priced, what was implied and what was accepted informally after the original agreement.

2. What work does that customer actually trigger? Sample completed orders or service events, not an org chart. Trace each through quotation, engineering, release, purchasing, production, delivery, acceptance, invoicing and correction. Measure first-pass acceptance and recurring exceptions.

3. Which requirements are necessary, valuable and paid for? Separate customer-requested differentiation, safety and quality obligations, deliberate reserves and true operating requirements from internal re-entry, repeat approvals, rescheduling and reconciliation. Test at least one plausible competing explanation.

4. Where is the work being repeated—and why? Locate the precise handoff, decision rule, data mismatch or commitment that generates repeat effort. Compare similar transactions and ask what evidence would disprove the suspected structural cause.

5. Could the same promise be fulfilled another way? Compare an alternative route, release standard, schedule, service tier, customer agreement or operating boundary. Preserve customer acceptance, product quality and required controls. Include transition costs and any new risks the alternative creates.

6. What are the verified economics? Reconcile revenue with direct and customer-specific costs, relevant capacity, incremental capital and working capital. Keep realized savings, paid time and contribution opportunities separate. Show which results were measured and which remain hypotheses.

7. What is the decision? Choose whether to expand the relationship, change the commercial terms, redesign the way the promise is fulfilled, limit or exit particular work—or deliberately change nothing. Specify the evidence, owner, measure and review date for the decision.

The test does not require that every demanding customer be retained, repriced or exited. It may conclude the customer is valuable, the service is correctly priced, and the apparent inefficiency is a necessary operating obligation. It may conclude that the customer should pay more. Or it may find that redesigning the company's own arrangement changes the economics while preserving the relationship.

The objective is not to prove that a difficult customer is unprofitable. It is to distinguish what the customer needs from what the company makes unnecessarily difficult.

8. What would change my conclusion?

I would weaken a Structural Drag hypothesis if sampled transactions show that the extra work is required by the customer's specification, legitimate compliance, quality, geography, actual demand variability or a carefully priced service promise—and that the current method performs it efficiently.

An allocated fixed cost that would remain, a profitable adjacent product mix, or deliberate ramp-up spending could also explain apparent poor contribution.

I would strengthen the case for structural examination if the same ordinary requests repeatedly cross functions incomplete, if one customer-specific exception is re-approved on every transaction, if different teams maintain conflicting records of a single promise, or if rerouting equivalent work reduces corrections while preserving quality and service.

A redesign that shifts the same effort to a supplier or customer, compromises service, or counts released hours twice has not established recoverable value.

9. Before winning the next big account

A major new account may be exactly what the company needs. Before the next $10 million in revenue arrives, management should know whether that work will enter a functioning operating path or create recurring exceptions.

The right response may be better pricing, a different route, a narrower promise, more capacity—or less repeated work.

Not every dollar of revenue deserves the same operating capacity.

Decide what work to accept, how to price it and how to fulfill it.

Research 01 examined growth without operating leverage. Research 02 examined technology investment. Research 03 examined cash conversion. Research 04 examined usable capacity and delivered output. Research 05 examined acquisition value. Research 06 asks whether the customer relationship and the way it is served create economic contribution.

Gower Idrees

Creator of the Business Structural Drag framework and the Renergy™ management framework

Sources, Evidence Notes & Research Methodology

Primary company documents reviewed through November 3, 2026. Public figures relate to the reporting periods specified; they do not imply that events or conditions continued unchanged after those periods.